Cobalt chloride stimulates phosphoinositide 3-kinase / Akt signaling through the epidermal growth factor receptor in oral squamous cell carcinoma

Tumor cells are often found under hypoxic conditions due to the rapid outgrowth of their vascular supply, and, in order to survive hypoxia, these cells induce numerous signaling factors. Akt is an important kinase in cell survival, and its activity is regulated by the upstream phosphoinositide 3-kinase (PI3K) and receptor tyrosine kinases (RTKs). In this study, we examined Akt activation and RTKs/PI3K/Akt signaling using the hypoxia-mimetic cobalt chloride in oral squamous carcinoma cells. Cobalt chloride increases Akt phosphorylation in both a doseand time-dependent manner. Blocking the activation of the PI3K/Akt pathway using LY294002 abolished Akt activation in response to cobalt chloride, suggesting that Akt phosphorylation by cobalt chloride is dependent on PI3K. In addition, activation of the PI3K/Akt pathway seems to rely on the epidermal growth factor receptor (EGFR), since the inhibition of EGFR attenuated cobalt chloride-induced Akt activation. The results in this study also demonstrate that cobalt chloride increases EGFR protein levels and induces oral squamous cell carcinoma cells to enter S phase. BIOCELL 2010, VOL. 34(1): 15-21 ISSN 0327 9545 PRINTED IN ARGENTINA


Introduction
Hypoxia, defined as a state where oxygen tension drops below normal limits, plays a pivotal role in pathological conditions, including tumors.Tumors, especially those that are cancerous, tend to be hypoxic because they grow too fast to contain enough blood vessels to maintain their oxygen supply.Hypoxia can limit tumor growth and tumors with poor vascularization fail to grow, leading to apoptosis.In contrast, certain types of cancer cells can survive and even proliferate under hypoxic conditions by triggering specific cellular and systemic adaptive responses (Hockel and Vaupel, 2001;Semenza, 2002).
When cells are exposed to hypoxia, a large number of factors respond to the altered environment, resulting in either cell death or survival.Of these factors, Akt is one of the most important signaling molecules for determining cell fate under hypoxic conditions.A few studies have reported that hypoxia or the hypoxia-mimicking condition produced by cobalt chloride (CoCl 2 ) treatment can activate Akt, whereas other investigations have suggested that it inhibits, or has no effect on Akt activation (Barthel et al., 2007;Chen et al., 2001;Lee et al., 2008a, b).Thus, Akt activity in response to hypoxia remains controversial.
Akt activity is mainly dependent on phosphoinositide 3-kinase (PI3K).Active PI3K phosphorylates membrane phosphatidylinositol phospholipids, and these lipid products act as second messengers that mediate the diverse cellular functions of PI3K.One of the targets of these lipid second messengers is the serine/threonine kinase Akt (Franke et al., 1997;Toker and Cantley, 1997).The PI3K/Akt pathway is activated by receptor tyrosine kinases (RTKs), such as the epidermal growth factor receptor (EGFR), the insulin growth factor 1 receptor (IGF1R), and the platelet-derived growth factor receptor (PDGFR).The RTKs are stimulated by their respective growth factors, leading to cell metabolism and growth through downstream signaling systems (Alessi et al., 1996;Blume-Jensen and Hunter, 2001;Chen et al., 2001).In addition to growth factor ligands, it has been suggested that heat shock and ultraviolet light can activate RTKs, resulting in the activation of PI3K and its target Akt (Coffer et al., 1995;Lin et al., 1997).Though numerous RTK candidates and the consecutive PI3K/Akt signaling pathways have been studied, there are few reports on how hypoxia affects RTKs/PI3K/Akt signaling.
This study examined the effect of hypoxia on signaling pathways in oral squamous cell carcinoma cells using CoCl 2 to induce a hypoxia-mimicking condition.First, we investigated whether CoCl 2 can activate or inhibit the PI3K/Akt pathway in oral squamous carcinoma cells.Second, we studied which growth factor signaling pathway is used when the stimuli evoke cellular responses.Finally, we investigated how the hypoxia-mimetic CoCl 2 affects the cell cycle in oral squamous carcinoma cells.

Cell culture
An YD8 cell line was established from the cancer of an oral squamous cell carcinoma patient and was gifted from the Department of Oral Pathology, College of Dentistry, Yonsei University (Seoul, Korea).The cells were grown in RPMI 1640 medium (Gibco, Rockville, MD) with 10% fetal bovine serum (Gibco) and incubated in a humidified atmosphere containing 5% CO 2 at 37ºC.

Western blot analysis
YD8 cells were lysed in lysis buffer (150 mM NaCl, 0.2% Nonidet P-40, 50 mM NaF, 100 μM Na 3 VO 4 , pH 7.2) in the presence of protease inhibitors (100 μM phenylmethylsulfonyl fluoride, 1 μg/ml leupeptin, 1 μg/ ml aprotinin) on ice for 30 min, followed by clarification by centrifugation at 12,000 g for 15 min.Protein concentrations were determined using the Bio-Rad Protein assay (Bio-Rad, Hercules, CA).Proteins (50 μg) were resolved on 10% (w/v) polyacrylamide gels and transferred onto polyvinylidene fluoride membranes (Bio-Rad).Membranes were blocked with 5% non-fat milk in Trisbuffered saline (TBS, pH 7.6) containing 0.1% Tween20 for 1 h at room temperature.Subsequently, the membranes were incubated overnight with the appropriate primary antibodies at 4ºC.The membranes were washed in Tris-buffered saline/Tween20 and were incubated with the respective horseradish peroxidase-conjugated secondary antibody in Tris-buffered saline/Tween20.After washing in Tris-buffered saline/Tween20, bound secondary antibodies were detected using Supersignal West Femto substrate (Pierce, Madison, WI).The immunoreactive bands were captured and quantified using a ChemiDoc system (Bio-Rad).All the experiments were done at least three times independently.

Cell cycle analysis
Cells were washed with phosphate buffered saline containing 1% bovine serum albumin and centrifuged at 2,000 rpm for 10 min.Cells were then washed with phosphate buffered saline and resuspended in 3 ml of ice-cold 75% ethanol for 24-72 h.The cells were incubated with DNase-free RNase (0.1 mg/ml; Sigma) and propidium iodide (50 μg/ml; Sigma) at 4ºC for at least 1 h in a dark room prior to flow cytometric analysis.The percentages of cells in different phases of the cell cycle were measured with a flow cytometer (Beckman Coulter, Fullerton, CA).A minimum of 10 5 cells were measured and analyzed for cell cycle analysis.

Statistical analysis
Results are presented as mean ± standard deviation of separate experiments, and were expressed as a ratio of treated cells to control cells.ANOVA or Student ttest was used for statistical analysis using a software package (SPSS 14.0).

A hypoxic condition induced by CoCl 2 increases Akt phosphorylation in oral squamous carcinoma cells
Akt activation by hypoxia was studied using CoCl 2, a hypoxia-mimicking agent.YD8 cells were treated with 0, 10, 50, 100, 200, or 300 μM of CoCl 2 for 3 h.The activation of Akt was assessed via immunoblotting us-ing specific antibodies raised against threonine (Thr) 308 and serine (Ser) 473 of phospho-Akt.These antibodies were chosen because full activation of Akt requires the phosphorylation of both residues (Alessi et al., 1996).Treatment with CoCl 2 resulted in an appreciable increase in detectable Thr308 and Ser473 phosphorylation without affecting total Akt expression in YD8 cells (Fig. 1).The increases of Akt phosphorylation at both residues were statistically significant (P<0.01).Next, to determine the effect CoCl 2 has on Akt activation over a time course, cells were treated with 300 μM of CoCl 2 for 1-24 h.As shown in figure 2, a significant increase in the level of phosphorylated Akt at Ser473 was observed 3 h after CoCl 2 treatment.This increase peaked at 6 h and returned to a basal Akt phosphorylation level at 24 h.However, phosphorylation at Thr308 was significantly higher in CoCl 2 -treated cells for 24 h, though the expression of the protein was weakened.These observations reveal that the hypoxia-mimetic CoCl 2 increases Akt phosphorylation at both residues in a dose-and time-dependent manner and that Thr308 is more sensitive to phosphorylation than Ser473 in the presence of CoCl 2 .

CoCl 2 activates Akt in a PI3K-dependent manner and through EGFR
We also attempted to investigate the molecular mechanism by which CoCl 2 induces Akt activation in oral squamous carcinoma cells.The level of Akt phosphorylation reflects PI3K activity, since Akt phosphorylation is invariably stimulated by active PI3K (Cantley, 2002;Toker and Cantley, 1997).To confirm the involvement of PI3K in CoCl 2 -induced Akt phosphorylation, we pretreated the cells with 20 μM of LY294002, a PI3K inhibitor, for 1 h and then added 300 μM of CoCl 2 to the cells for 3 h.CoCl 2 -stimulated Akt phosphorylation at Thr308 and Ser473 was significantly inhibited in oral squamous carcinoma cells in the presence of LY294002, suggesting that the phosphorylation of Akt is PI3K-dependent (Fig. 3).
Increased Akt phosphorylation by CoCl 2 in this study indicates that the hypoxic condition provides survival signals to cells, such as growth factors.To clarify the involvement of growth factor receptors and to define which growth factor signaling pathways contribute to the activation of the PI3K/Akt pathway by CoCl 2 , we investigated how blocking individual RTKs affects Akt activation in cells exposed to CoCl 2 .We used 1 μM PD168393, 2 nM picropodophyllin (PPP), and 200 nM DMPQ as specific inhibitors of EGFR, IGF1R, and PDGFR, respectively.Though picropodophyllin showed inhibitory activity of Akt phosphorylation, it had no inhibitory effect on Akt phosphorylation induced by CoCl 2 .CoCl 2 -induced Akt phosphorylation was suppressed only by PD168393, but not by picropodophyllin or DMPQ, implying that hypoxic signaling is mediated by EGFR (Fig. 4).

CoCl 2 treatment results in the upregulation of EGFR and an increase in DNA synthesis
To investigate whether CoCl 2 directly affects EGFR itself, the level of EGFR expression was evaluated in YD8 cells in the presence of varying CoCl 2 concentrations.EGFR protein levels were increased by CoCl 2 in a dose-dependent manner.When cells were treated with 300 μM of CoCl 2 for 24 h, the increase in EGFR expression was comparably prominent (Fig. 5).In addi- tion, the effect of CoCl 2 on the cell cycle was examined using a flow cytometer.CoCl 2 increased the percentage of the cell population in the S-phase (45.6%) compared to control-treated cells (21.9%) (Fig. 6).FIGURE 6. Signaling by the hypoxia-mimetic CoCl 2 results in S-phase entry.Cells were incubated with 300 μM of CoCl 2 .After 1 h, 3 h, 6 h, and 24 h, cells were harvested, fixed in ethanol, and stained with propidium iodide.The cell cycle was analyzed by flow cytometry.The number of S (%) depicts the percentage of S-phase fraction of control or CoCl 2treated cells.

Discussion
Activation of the PI3K/Akt pathway drives tumors towards a more malignant phenotype, which can survive under harsh conditions (Nicholson and Anderson, 2002).A study by Izuishi et al. (2000) investigated the role of Akt in cell tolerance to nutrient deprivation, and found that high cellular Akt expression was associated with a higher tolerance of liver and pancreatic cancer cell lines to nutrient deprivation.The present study shows that CoCl 2 , a well-known hypoxia-mimetic agent, activates Akt in oral squamous carcinoma cells, which concurs with previous reports using Hela and PC12 cells (Alvarez-Tejado et al., 2001;Jin et al., 2007).In contrast to our data and the data observed in Hela and PC12 cells, numerous studies have reported that hypoxia or cobalt ions have no effect on Akt phosphorylation and even inhibit Akt phosphorylation in certain cells, including smooth muscle cells, gastric carcinoma cells, and COS cells (Ardyanto et al., Humar et al., 2002;Loberg et al., 2002).Finally, it has been proposed that the activation of Akt during hypoxia may be celltype dependent and not a general effect of hypoxia (Alvarez-Tejado et al., 2001).However, the differences in Akt activation by hypoxia or CoCl 2 cannot be explained solely by the cell type because Akt phosphorylation results differed in previous reports using the same cell line.For example, Jin et al. reported that Akt was activated in Hela cells in response to CoCl 2 exposure, whereas Barthel et al. observed that CoCl 2 had no effect on Akt activation in these cells, though other metal ions, such as copper and zinc, significantly induced Akt phosphorylation (Barthel et al., 2007;Jin et al., 2007).The main experimental difference between these two studies was the CoCl 2 incubation time.Barthel et al. (2007) investigated Akt phosphorylation after only a 30 min exposure to CoCl 2 , while Jin et al. (2007) observed Akt activation after 6 h of CoCl 2 treatment, similar to the present study.In a report using PC12 cells, Alvarez-Tejado et al. (2001) suggested that Akt is activated several hours after exposure to a hypoxic environment, whereas growth factors stimulate Akt within minutes (Alvarez-Tejado et al., 2001).The authors concluded that Akt activation by hypoxia or the hypoxia-mimetic CoCl 2 may depend not only on the cell type, but also on the exposure time to these conditions.Furthermore, the hypoxic condition requires several hours to induce its effects on the Akt pathway, though short-period hypoxia clearly activates Akt.
Although IGF1R and PDGFR are well-known PI3K/Akt stimulators, the presence of inhibitors specifically targeting IGF1R or PDGFR does not block CoCl 2 -induced Akt activation in oral squamous carcinoma cells.The results of the present study showed that CoCl 2 induces the activation of PI3K/Akt and that this occurs through an EGFR-related pathway, but unlikely through IGF1R or PDGFR.In addition to the inhibition of CoCl 2 -induced Akt activation by the EGFR inhibitor, the upregulation of EGFR by CoCl 2 was also observed in this study.A recent study using oligonucle-otide array analyses showed that hypoxia induced EGFR as well as other well-known hypoxia-induced genes, such as VEGF.Using immunohistochemistry, this study observed the induction of EGFR in tumor cells in the vicinity of necrotic areas, which is a histological indicator of tumor hypoxia (Wang et al., 2007).Also, it has been suggested that the hypoxic microenvironment induces the overexpression of EGFR via HIF-2α activation (Franovic et al., 2007).Taken together, these results suggest that oral squamous carcinoma cells use the EGFR signaling pathway to adapt and survive in a hypoxic environment.
It has been reported that Akt plays an important role in the transition from the G1 phase to the DNA synthesis phase of the cell cycle.Lee et al. (2008a) have suggested that short periods of hypoxia increase the percentage of the cell population in the S phase by inducing cell cycle regulatory proteins through the PI3K/ Akt/mTOR pathways in mouse embryonic cells.It was also previously reported that a short exposure time to hypoxia increases DNA synthesis in primary cultured chicken hepatocytes (Lee et al., 2008b).While currently no such results have been reported for human cells, it seems that exposure to hypoxia increases DNA synthesis in oral squamous carcinoma cells.This transition to the S phase may allow cancer cells to survive and become resistant to radiotherapy and/or chemotherapy.
In conclusion, this study demonstrates that CoCl 2induced hypoxia in oral squamous carcinoma cells increases Akt activation through EGFR signaling, causing the cells to enter S phase.We propose that oral squamous carcinoma cells are resistant to hypoxia and that under this condition, EGFR expression is induced, resulting in the activation of Akt.This Akt induction may promote cell cycle progression and be an important factor in resistance to chemotherapy and/or radiotherapy in hypoxic oral squamous carcinoma cells.

FIGURE 2 .
FIGURE 2. The effects of CoCl 2 treatment on Akt phosphorylation.YD8 cells were treated with 300 μM of CoCl 2 for the indicated times.Then, the expression of phospho-Akt (Thr308) and phospho-Akt (Ser473) was analyzed by western blot analysis.Quantitative data of densitometric analysis were presented in the histograms at the bottom of figures.The data are presented as a ratio of the control, set at 1. * and ** indicate a statistical significance of p <0.05 and P<0.01, respectively, as determined by paired Student t-test.

FIGURE 1 .
FIGURE 1.The effects of CoCl 2 treatment on Akt activation.CoCl 2 increases in a dose-dependent fashion.YD8 cells were treated with varying concentrations of CoCl 2 as indicated for 3 h, and the expression of phospho-Akt (Thr308) and phospho-Akt (Ser473) implying Akt activation were analyzed by western blot analysis.The density of the bands for phosphorylated Akt was quantitated by densitometric analysis and data are expressed as a ratio of the control, set at 1.The error bars indicate the standard deviation of the three separate experiments.The increase of phosphorylated Akt protein by CoCl 2 treatment was statistically significant (P<0.05;ANOVA analysis).

FIGURE 3 .
FIGURE 3. The effects of a PI3K inhibitor on the phosphorylation of Akt in cells exposed to CoCl 2 .Cells were pretreated with the PI3K inhibitor for 1 h and then exposed to 300 μM of CoCl 2 .Following 3 h incubation, lysates were subjected to western blot analysis for the detection of phosphorylated Akt proteins.The density of the bands was quantified by densitometric analysis as reported in the histograms at the bottom of the figure.Data are expressed as a ratio of the non-treated control, set at 1.The PI3K inhibitor (LY; 20 μM LY294002) significantly suppressed CoCl 2 -induced Akt phosphorylation (*, p <0.01; **, P<0.05; paired Student t-test).

FIGURE 4 .
FIGURE 4. The effects of various kinase inhibitors on the phosphorylation of Akt in cells exposed to CoCl 2 .Cells were treated with the kinase inhibitors for 1 h and then exposed to 300 μM of CoCl 2 for 3 h.The expression of phospho-Akt (Thr308) and phospho-Akt (Ser473) was analyzed by western blot analysis.The density of the bands was quantified by densitometric analysis.The effect of kinase inhibitors on Akt phosphorylation was expressed relative to that of untreated control cells, set at 1.In the histograms data are presented as mean ± standard deviation.PD, 1 μM PD168393 (EGFR inhibitor); PPP, 2 nM picropodophyllin (IGF1R inhibitor); DMPQ, 200 nM DMPQ (PDGFR inhibitor).The effects of inhibitors on CoCl 2 -induced Akt phosphorylation was examined and analyzed (**, p <0.01; paired Student t-test).

FIGURE 5 .
FIGURE 5.The effect of CoCl 2 treatment on EGFR expression.CoCl 2 treatment increases the expression of EGFR protein in a dose-and time-dependent manner.(A) YD8 cells were treated with varying concentrations of CoCl 2 as indicated.After 3 h, the cells were analyzed for EGFR.The expression of EGFR protein was significantly increased by CoCl 2 treatment (P< 0.05; ANOVA analysis).(B) YD8 cells were treated with 300 μM of CoCl 2 for 1-24 h.Cells were harvested and analyzed for Akt phosphorylation.The density of bands was quantitated and EGFR levels were expressed relative to that of untreated control cells, set at 1 (*, p <0.01; **, P<0.05; paired Student t-test).